Introduction to R for Data Analysis

# Introduction to R for Data Analysis
## Data Visualization - Part 1
### Johannes Breuer & Stefan Jünger
### 2021-08-03

---

---

## Why should we use data visualization?
While we know that all of you are familiar with the concept of data visualization, we want to briefly discuss why we think it's essential to be familiar with and use it.

In general:
- Good plots can contribute to a better understanding of your analysis results
- Plots also help you to understand your data in the first place

- Generating a plot is easy as you will see
- ... Making good plots, however, can take a while

---

## Plots in `R`
- `R` is fun, and so is creating plots in `R`

- Almost every plot type is supported in `R`
- either in your standard installation or in through additional packages

- A large number of exports formats are supported
- `.png`, `.jpg`, `.tiff`, `.svg`, `.bmp`, `.pdf`, `.eps`, etc.

---

## We'll start rather basic

---

## Content of the visualization sessions
.pull-left[
**`Base R` visualization**
- Standard plotting procedures in R
- very short
]

.pull-right[
**`tidyverse`/`ggplot2` visualization**
- Modern interface to graphics
- grammar of graphics
]

There's more that we won't cover:
- [`lattice`](https://cran.r-project.org/web/packages/lattice/index.html) plots, for example

---

## Graphics in `R`
Since the early days, graphics are a first-class citizen in `R`.

A standard `R` installation doesn't require any additional packages to create graphics. It's part of the `graphics` package.

```r
barplot(table(gp_covid$hzcy001a))
```
]

]

---

## Ok, but let's start from the beginning
The most basic function to plot in R is `plot()`.

```r
plot(gp_covid$hzcy001a)
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-3-1.png" style="display: block; margin: auto;" />
]

---

## We can turn this into a bivariate scatterplot
.pull-left[

```r
plot(
  gp_covid$age_cat, 
  gp_covid$hzcy001a
)
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-4-1.png" style="display: block; margin: auto;" />
]

---

## Add some jitter and also change the points type
.pull-left[

```r
plot(
  jitter(gp_covid$age_cat, 2), 
  jitter(gp_covid$hzcy001a, 2),
  pch = 16
)
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-5-1.png" style="display: block; margin: auto;" />
]

---

## Adding stuff to the plot: titles & labels
.pull-left[

```r
plot(
  jitter(gp_covid$age_cat, 2), 
  jitter(gp_covid$hzcy001a, 2),
  pch = 16,
  main = "Relationship between Age and Subjective Risk of a COVID-19 Infection",
  xlab = "Age of Respondents",
  ylab = "Subjective Risk of Being Infected"
)
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-6-1.png" style="display: block; margin: auto;" />
]

---

## Adding stuff to the plot: axis labels
.tinyish[
.pull-left[

```r
plot(
  jitter(gp_covid$age_cat, 2), 
  jitter(gp_covid$hzcy001a, 2),
  pch = 16,
  main = "Relationship between Age and Risk of Covid-19 Infection",
  xlab = "Age of Respondents",
  ylab = "Subjective Risk of Being Infected",
  yaxt = "n"
)
axis(
  side = 2, 
  at = 1:7,
  labels = c(
    "Not at all",
    "Very\nunlikely", 
    "Rather\nunlikely",
    "Moderately",
    "Rather", 
    "Very",
    "Absolutely"
  ),
  las = 0
)
```
]
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-7-1.png" style="display: block; margin: auto;" />
]

---

# [Exercise](https://jobreu.github.io/r-intro-gesis-2021/exercises/Exercise_3_2_1_A_Simple_Plot.html) time 🏋️‍♀️💪🏃🚴

## [Solutions](https://jobreu.github.io/r-intro-gesis-2021/solutions/Exercise_3_2_1_A_Simple_Plot.html)

---

## Record your plot
Adding more and more elements to your plot also means that the code you have to write gets more and more. But what can we do when we want to re-use the same plot and dynamically add some stuff? We can record the plot!

```r
plot(
  jitter(gp_covid$age_cat, 2), 
  jitter(gp_covid$hzcy001a, 2),
  pch = 16,
  main = "Relationship between Age and Subjective Risk of Covid-19 Infection",
  xlab = "Age of Respondents",
  ylab = "Subjective Risk of Being Infected",
  yaxt = "n"
)

my_scatterplot <- recordPlot()
```
]
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-8-1.png" style="display: block; margin: auto;" />
]

---

## Now apply it to the previous plot

```r
my_scatterplot
axis(
  side = 2, 
  at = 1:7,
  labels = c(
    "Not at all",
    "Very\nunlikely", 
    "Rather\nunlikely",
    "Moderately",
    "Rather", 
    "Very",
    "Absolutely"
  ),
  las = 0
)
```
]
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-9-1.png" style="display: block; margin: auto;" />
]

---

## If you're happy, just update your recording

```r
my_scatterplot
axis(
  side = 2, 
  at = 1:7,
  labels = c(
    "Not at all",
    "Very\nunlikely", 
    "Rather\nunlikely",
    "Moderately",
    "Rather", 
    "Very",
    "Absolutely"
  ),
  las = 0
)

my_scatterplot <- recordPlot()
```
]
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-10-1.png" style="display: block; margin: auto;" />
]

---

## Where to go from here in `Base R`'s graphics?

Using similar procedures, we can add more and more stuff to our plot or edit its elements:
- regression lines
- legends
- annotations
- colors

- etc.

We surely also apply different *plot types*, such as
- histograms
- barplots
- boxplots
- densities
- pie charts
- etc.

---

## Example: A simple boxplot

```r
boxplot(
  gp_covid$hzcy001a ~
    as.factor(gp_covid$age_cat)
)
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-11-1.png" style="display: block; margin: auto;" />
]

---

## My attempt on Trump's plot

```r
barplot(
  table(gp_covid$age_cat)[c(2, 9:7)], 
  col = 
    c(
      "#4F94CD",
      "#FFA54F",
      "#CD9B1D",
      "#FF82AB"
    ),
  horiz = TRUE,
  axes = FALSE
)
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-12-1.png" style="display: block; margin: auto;" />
]

---

## One last detail: the `par()` and `dev.off()` function
`par()` stands for graphical parameters and is called before the actual plotting function. It prepares the graphics device in `R` and tells it: "Hey, plot(s) incoming!" A lot can be done within this function. The most commonly used options are for "telling" the device that 2, 3, 4, or `x` plots have to be printed.

We have used the option `mfrow` before by telling it how many rows (the first value in the vector) and columns (the second value in the vector) we aim to plot.

```r
par(mfrow = c(2, 2))
```

One caveat of using this function is that we actively have to turn off the device before generating another independent plot.

```r
dev.off()
```

---

## Exporting graphics
It's nice that `R` provides such pleasant plotting opportunities.

However, to include them in our papers, we need to export them.

As said in the beginning, numerous export formats are available in `R`.

---

## Export with *RStudio*

---

## Saving graphics via a command
Alternatively, you can also export plots with the commands `png()`, `pdf()` or `jpeg()`, for example. For this purpose, you first have to wrap the plot call between one of those functions and a `dev.off()`call.

```r
png("Histogram.png")
plot(gp_covid$age_cat)
dev.off()
```

```r
pdf("Histogram.pdf")
plot(gp_covid$age_cat)
dev.off()
```

```r
jpeg("Histogram.jpeg")
plot(gp_covid$age_cat)
dev.off()
```

It's that easy.

---

# [Exercise](https://jobreu.github.io/r-intro-gesis-2021/exercises/Exercise_3_2_2_Handling_Multiple_Plots.html) time 🏋️‍♀️💪🏃🚴

## [Solutions](Exercise_3_2_2_Handling_Multiple_Plots.html)

---

## My personal note on `base R` plotting
Hopefully, you have gotten the feeling that the `base R` base techniques for plotting already are well-suited for your daily data exploration needs.

But to be honest: I do not use all the other functions that often. The syntax is sometimes cumbersome with all the `par()`  or `dev.off()` calls, and manipulating parameters simply feels somewhat "outdated".

Now, we will learn more modern techniques using `ggplot2`. Yet, we still believe that it is worthwhile to become comfortable with `base R` plotting since `ggplot2`, e.g., may sometimes be "too much" for simple data exploration.

**As so often, in the end, it's also a matter of taste.**

---

## What is `ggplot2`?
`ggplot2` is another `R` package for creating plots and is part of the `tidyverse`.

It uses the *grammar of graphics*. Some things to note about `ggplot2`:
- it is well-suited for multi-dimensional data
- it expects data (frames) as input
- components of the plot are added as layers

```r
plot_call +
  layer_1 +
  layer_2 +
  ... +
  layer_n
```

---
## `ggplot2` examples 
.pull-left[
<img src="data:image/png;base64,#C:\Users\mueller2\talks_presentations\r-intro-gesis-2021\content\img\143_radar_chart_multi_indiv_2.png" width="640" style="display: block; margin: auto;" />
]

.pull-right[
<img src="data:image/png;base64,#C:\Users\mueller2\talks_presentations\r-intro-gesis-2021\content\img\21_ggplot2_ddensity_plot.png" width="640" style="display: block; margin: auto;" />
]

Sources: https://www.r-graph-gallery.com/wp-content/uploads/2016/05/143_radar_chart_multi_indiv_2.png and https://www.r-graph-gallery.com/wp-content/uploads/2015/09/21_ggplot2_ddensity_plot.png

---

## `ggplot2` examples 
.pull-left[
<img src="data:image/png;base64,#C:\Users\mueller2\talks_presentations\r-intro-gesis-2021\content\img\51_scatterplot_linear_model_with_CI_ggplot2.png" width="640" style="display: block; margin: auto;" />
]

.pull-right[
<img src="data:image/png;base64,#C:\Users\mueller2\talks_presentations\r-intro-gesis-2021\content\img\328_Hexbin_map_USA_4.png" width="800" style="display: block; margin: auto;" />
]

Sources: https://www.r-graph-gallery.com/wp-content/uploads/2015/11/51_scatterplot_linear_model_with_CI_ggplot2-300x300.png and https://www.r-graph-gallery.com/wp-content/uploads/2017/12/328_Hexbin_map_USA_4-300x200.png

---

## Barplots as in `base R`

```r
ggplot(gp_covid, aes(x = age_cat)) + 
  geom_bar()
```
]
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-19-1.png" style="display: block; margin: auto;" />
]

---

## Boxplots as in `base R`

```r
ggplot(
  gp_covid, 
  aes(x = as.factor(age_cat), y = sum_trust)) + 
  geom_boxplot()
```
]
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-20-1.png" style="display: block; margin: auto;" />
]

---

## Components of a plot
According to Wickham (2010, 8)* a layered plot consists of the following components:

* http://dx.doi.org/10.1198/jcgs.2009.07098

- data and aesthetic mappings,
- geometric objects,
- scales,
- and facet specification

```r
plot_call +
  data +
  aesthetics +
  geometries +
  scales +
  facets
```

---

## Data requirements
You can use one single data frame to create a plot in `ggplot2`.
- everything on the plot is just data
- creates a smooth workflow from data wrangling to the final presentation of the results

Source: http://r4ds.had.co.nz

However, this makes it difficult to add extra features to your plot.
- There are ways of doing it anyway
- Yet, it requires thinking about what to plot

---

## Why the long format? 🐴
`ggplot2` prefers data in the long data format (**NB**: of course, only if this is possible and makes sense for the dataset at hand)
- in some scientific disciplines this format is only used for specialized analyses (e.g., time series analysis)

.pull-left[
We may want to get used to it as this format has some benefits:
- every element we aim to plot is an observation
- no thinking required how a specific variable relates to an observation
- most importantly, the long format is more parsimonious
- it requires less memory and less disk space
]

.pull-right[
<img src="data:image/png;base64,#C:\Users\mueller2\talks_presentations\r-intro-gesis-2021\content\img\long.png" width="40%" style="display: block; margin: auto;" />
Source: https://github.com/gadenbuie/tidyexplain#tidy-data
]

---

## Before we start
The architecture of building plots in `ggplot` is similar to standard `R` graphics. There is an initial plotting call, and subsequently, more stuff is added to the plot.

However, in `base R`, it is sometimes tricky to find out how to add (or remove) certain plot elements. For example, think of removing the axis ticks in the scatter plot.

We will systematically explore which elements are used in `ggplot` in this session.

---

## Scatterplot from earlier

```r
ggplot(
  data = gp_covid,
  aes(
    x = age_cat, 
    y = sum_trust
  )
) +
  geom_point()
```
]
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-21-1.png" style="display: block; margin: auto;" />
]

---

## Scatterplot from earlier with jitter

```r
ggplot(
  data = gp_covid,
  aes(
    x = age_cat, 
    y = sum_trust
  )
) +
  geom_jitter()
```
]
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-22-1.png" style="display: block; margin: auto;" />
]

---

## Creating your own plot
We do not want to give a lecture on the theory behind data visualization (if you want that, we suggest having a look at the excellent book [*Fundamentals of Data Visualization*](https://serialmentor.com/dataviz/) by Claus O. Wilke).
- creating plots is all about practice
- ...and 'borrowing' code from others

```r
please +
  no +
  more +
  pseudo +
  code +
  man
```
]

.column-right-half[
Three components are important:
- Plot initiation
- data input
- aesthetics definition
- so-called geoms
]

---

## Plot initiation
Now, let's start from the beginning and have a closer look at the *grammar of graphics*.

```r
ggplot()
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-23-1.png" style="display: block; margin: auto;" />
]

**But it doesn't show anything...**

---

## What now? Data input!

```r
ggplot(data = gp_covid)
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-24-1.png" style="display: block; margin: auto;" />
]

**Still nothing there...**

---

## `aes`thetics!

```r
ggplot(data = gp_covid) +
  aes(x = age_cat, y = sum_trust)
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-25-1.png" style="display: block; margin: auto;" />
]

**That's a little bit better, right?**

---

## `geom`s!
.pull-left[
Finally, `ggplot` needs information *how* to plot the variables.

```r
ggplot(data = gp_covid) +
  aes(x = age_cat, y = sum_trust) +
  geom_point()
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-26-1.png" style="display: block; margin: auto;" />
]

**A scatter plot!**

---

## Add a fancy `geom`
.pull-left[
We can also add more than one `geom`.

```r
ggplot(data = gp_covid) +
  aes(x = age_cat, y = sum_trust) +
  geom_jitter() +
  geom_smooth(method = "lm", se = FALSE)
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-27-1.png" style="display: block; margin: auto;" />
]

**A regression line!** (without confidence intervals; the regression behind this operation is run automatically)

---

# [Exercise](XXX) time 🏋️‍♀️💪🏃🚴

## [Solutions](XX)

---

## Goind further: Working with grouping variables

```r
gp_covid <-
 gp_covid %>% 
 sjlabelled::remove_all_labels() %>% 
 mutate(
 pol_leaning_cat = 
 case_when(
 between(political_orientation, 0, 3) ~ "left",
 between(political_orientation, 4, 7) ~ "center",
 political_orientation > 7 ~ "right"
 ) 
 ) %>% 
 filter(pol_leaning_cat != "NA")
```

---

## Going further: adding group `aes`thetics

```r
ggplot(
  data = gp_covid,
  aes(
    x = age_cat, 
    y = sum_trust, 
    group = pol_leaning_cat
  )
) +
  geom_smooth(method = "lm")
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-28-1.png" style="display: block; margin: auto;" />
]

---

## Manipulating group `aes`thetics

```r
ggplot(
  data = gp_covid,
  aes(
    x = age_cat, 
    y = sum_trust, 
    color = pol_leaning_cat,
    group = pol_leaning_cat
  )
) +
  geom_smooth(method = "lm", se = FALSE) 
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-29-1.png" style="display: block; margin: auto;" />
]

The legend is drawn automatically, that's handy!

---

## Using another palette

```r
ggplot(
  data = gp_covid,
  aes(
    x = age_cat, 
    y = sum_trust, 
    color = pol_leaning_cat,
    group = pol_leaning_cat
  )
) +
  geom_smooth(method = "lm", se = FALSE) +
  scale_color_brewer(palette = "Spectral") 
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-30-1.png" style="display: block; margin: auto;" />
]

---

## Difference between `color` and `fill`
When you work with `ggplot2`, at least after some time, you will be faced with two components of the plot or `geom` associated with colors: `color` and `fill`.

Generally, `color` refers to the geometry borders, such as a line. `fill` refers to a geometry area, such as a polygon.

Have this difference in mind when you use `scale_color_brewer` or `scale_fill_brewer` in your plots. Manipulating these colors and their corresponding legends in an elaborate plot can get really tedious, to be honest.

---

## Choosing a fill color

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-31-1.png" style="display: block; margin: auto;" />
]

---

## Colors and `theme`s
One particular strength of `ggplot2` lies in its immense theming capabilities. The package has some built-in theme functions that makes theming a plot fairly easy, e.g.,
- `theme_bw()`
- `theme_dark()`
- `theme_void()`
- etc.

See: https://ggplot2.tidyverse.org/reference/ggtheme.html

If you want to, you can play around with some of those themes in the exercises for this session.

In general, the [`r-color-palettes` repository by Emil Hvitfeldt](https://github.com/EmilHvitfeldt/r-color-palettes) is a good resource for choosing color palettes in `R` and there are many collections of full `ggplot2` themes out there (e.g., the [`hrbrthemes` package](https://github.com/hrbrmstr/hrbrthemes)).

---

## Alternative to being too colorful: facets
.pull-left[

```r
ggplot(
  data = gp_covid,
  aes(
    x = age_cat, 
    y = sum_trust
  )
) +
  geom_smooth(color = "black", method = "lm", se = FALSE) +
  facet_wrap(~pol_leaning_cat, ncol = 3) +
  theme_bw()
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-32-1.png" style="display: block; margin: auto;" />
]

---

## The `theme()` argument in general
The most direct interface for manipulating your theme is the `theme()` argument. Here you can change the appearance of:
- axis labels
- captions and titles
- legend
- grid layout
- the wrapping strips
- ...

---

## Example: changing the grid layout & axis labels
.pull-left[

```r
ggplot(
  data = gp_covid,
  aes(
    x = age_cat, 
    y = sum_trust
  )
) +
  geom_smooth(color = "black", method = "lm", se = FALSE) +
  facet_wrap(~pol_leaning_cat, ncol = 3) +
  theme_bw() +
  theme(
    panel.grid.major = element_blank(), 
    panel.grid.minor = element_blank(),
    strip.background = element_rect(fill = "white")
  )
```
]

.pull-right[ 
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-33-1.png" style="display: block; margin: auto;" />
]

---

## Example: changing axis labels
.pull-left[

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-34-1.png" style="display: block; margin: auto;" />
]

---

## Another remark on plotting options

Particularly, plotting similar aesthetics which interfere with the automatic procedures can create conflicts.

Some 'favourites' include:
- Multiple legends
- and various color scales for similar `geoms` 
- ... and there's more!
]

.pull-right[
<img src="data:image/png;base64,#C:\Users\mueller2\talks_presentations\r-intro-gesis-2021\content\img\800px-The_Scream.jpg" width="1065" style="display: block; margin: auto;" />
]

---

## `ggplot` plots are 'simple' objects
In contrast to standard `R` plots, `ggplot2` are standard objects like any other object in `R` (they are lists). So there is no graphics device involved from which we have recorded our plot to re-use it later. We can just use it directly.

```r
my_fancy_plot <- 
 ggplot(data = gapminder_children) +
 aes(x = year, y = children) +
 geom_point()

my_fancy_plot <- 
 my_fancy_plot +
 geom_smooth()
```

Additionally, there is also no need to call `dev.off()`

---

## It makes combining plots easy
As of today, there are now a lot of packages that help to combine `ggplot2`s fairly easily. For example, the [`cowplot` package](https://cran.r-project.org/web/packages/cowplot/index.html) provides a really flexible framework. I have used `cowplot` to create the map in the previous session.

Yet, fiddling with this package can become quite complicated. One of my favorite packages at the moment is the [`patchwork` package](https://cran.r-project.org/web/packages/patchwork/index.html) because of its easy to use syntax. It's really, really easy.

---

## Plotting side by side in one row
.pull-left[

```r
library(patchwork)

my_barplot <- 
 ggplot(
 gp_covid, 
 aes(x = age_cat)
 ) + 
 geom_bar()

my_boxplot <- 
 ggplot(
 gp_covid, 
 aes(y = age_cat)
 ) + 
 geom_boxplot()

my_barplot | my_boxplot
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-36-1.png" style="display: block; margin: auto;" />
]

---

## Plotting in two columns
.pull-left[

```r
my_barplot / my_boxplot
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-37-1.png" style="display: block; margin: auto;" />
]

---

## combine them with base R graphics
.pull-left[

```r
(my_barplot | ~barplot(table(gp_covid$age_cat))) /
  (my_boxplot | ~boxplot(gp_covid$age_cat))
```
]

.pull-right[
<img src="data:image/png;base64,#3_2_Data_Visualization_Part_1_files/figure-html/unnamed-chunk-38-1.png" style="display: block; margin: auto;" />
]

---

## There's more
You can also annotate plots with titles, subtitles, captions, and tags.

You can nest plots and introduce more complex layouts.

I'd recommend, if you're interested, to check out the [`patchwork` repository on *GitHub*](https://github.com/thomasp85/patchwork) since everything is really well-documented there.

---

## Exporting ggplot graphics
Exporting `ggplot2` graphics is fairly easy with the `ggsave()` function. It automatically detects the file format. You can also define the plot height, width, and dpi, which is particularly useful to produce high-class graphics for publications.

```r
nice_plot <- 
 qplot(
 x = gapminder_children$year, 
 y = gapminder_children$children,
 geom = c("point", "smooth")
 )

ggsave("nice_plot.png", nice_plot, dpi = 300)
```

Or:

```r
ggsave("nice_plot.tiff", nice_plot, dpi = 300)
```

---

## Now, what about visualizing explorative statistics?

Here are a few examples.

---

## Plotting structure of missing data

```r
library(visdat)

vis_miss(gp_covid[,1:20])
```

---

## Fancier barplots I

```r
library(scales)

gp_covid %>% 
  ggplot(aes(x = education_cat, 
             fill = education_cat)) + 
  geom_bar(aes(y = (..count..)/sum(..count..))) +
  scale_y_continuous(labels = scales::percent) +
  ylab("Relative Frequencies")
```

---

## Fancier barplots II

```r
gp_covid %>% 
  filter(!is.na(choice_of_party)) %>% 
  ggplot(aes(x = choice_of_party, 
             fill = choice_of_party)) + 
  geom_bar(aes(y = (..count..)/sum(..count..))) +
  scale_y_continuous(labels=scales::percent) +
  ylab("Relative Frequencies")
```

---

## Correlation plots

```r
library(GGally)

gp_covid %>% 
  select(hzcy044a:hzcy052a) %>% 
  ggcorr(
    label = TRUE,
    label_round = 2
  )
```

---

## Some additional resources

- [ggplot2 - Elegant Graphics for Data Analysis](https://www.springer.com/gp/book/9783319242750) by Hadley Wickham

- [Chapter 3](https://r4ds.had.co.nz/data-visualisation.html) in *R for Data Science*

- [Fundamentals of Data Visualization](https://serialmentor.com/dataviz/) by Claus O. Wilke

- [Data Visualization - A Practical Introduction](https://press.princeton.edu/titles/13826.html) by Kieran Healy

- [data-to-viz](https://www.data-to-viz.com/)

- [R Graph Gallery](https://www.r-graph-gallery.com/)

- [BBC Visual and Data Journalism cookbook for R graphics](https://bbc.github.io/rcookbook/#how_to_create_bbc_style_graphics)

- [ggplot2 extensions](http://www.ggplot2-exts.org/gallery/)

---

# [Exercise](XXX) time 🏋️‍♀️💪🏃🚴

## [Solutions](XXX)